Virtual file system

• US 20050114350A1
• Filed: 11/30/2004
• Published: 05/26/2005
• Est. Priority Date: 11/28/2001
• Status: Active Grant

• Current Assignee: Interactive Content Engines, LLC
• Original Assignee: Interactive Content Engines, LLC
• Inventors: Rose, Steven W., Rhoads, Neil A., Abdul, Corinna G.
• Granted Patent: US 7,644,136 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/0643 Management of files
• CPC Class Codes: G06F 3/0667 at data level, e.g. file, record or object virtualisation
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: H04L 65/1101 Session protocols
• CPC Class Codes: H04L 65/612 for unicast
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: H04N 21/2182 involving memory arrays, e.g. RAID disk arrays RAID arrays per se G06F3/0689; use of parity to protect data in RAID systems G06F11/1008
• CPC Class Codes: H04N 21/21825 involving removable storage units, e.g. tertiary storage such as magnetic tapes or optical disks
• CPC Class Codes: H04N 21/2318 using striping
• CPC Class Codes: H04N 21/2405 Monitoring of the internal components or processes of the server, e.g. server load allocation of resources in multiprogramming arrangements G06F9/50; performance measurement of computer activity G06F11/34
• CPC Class Codes: H04N 7/17318 Direct or substantially direct transmission and handling of requests

Abstract

A virtual file system including multiple storage processor nodes including a management node, a backbone switch, a disk drive array, and a virtual file manager executing on the management node. The backbone switch enables communication between the storage processor nodes. The disk drive array is coupled to and distributed across the storage processor nodes and stores multiple titles. Each title is divided into data subchunks which are distributed across the disk drive array in which each subchunk is stored on a disk drive of the disk drive array. The virtual file manager manages storage and access of each subchunk, and manages multiple directory entries including a directory entry for each title. Each directory entry is a list of subchunk location entries in which each subchunk location entry includes a storage processor node identifier, a disk drive identifier, and a logical address for locating and accessing each subchunk of each title.

Claims

1. A virtual file system, comprising:

a plurality of storage processor nodes including at least one management node;

a backbone switch, coupled to said plurality of storage processor nodes, that enables communication between each of said plurality of storage processor nodes;

a disk drive array coupled to and distributed across said plurality of storage processor nodes, said disk drive array storing a plurality of titles, each title divided into a plurality of subchunks which are distributed across said disk drive array in which each subchunk is stored on a disk drive of said disk drive array; and

said at least one management node executing a virtual file manager which manages storage and access of each subchunk of said plurality of titles, and which maintains a plurality of directory entries including a directory entry for each title, each said directory entry comprising a list of subchunk location entries in which each subchunk location entry comprises a storage processor node identifier, a disk drive identifier, and a logical address for locating and accessing each subchunk of each title stored on said disk drive array.

2. The virtual file system of claim 1, wherein each of said plurality of subchunks is retrieved in a single seek operation by providing said logical address to an identified disk drive of an identified storage processor node.

3. The virtual file system of claim 1, wherein the full capacity of each disk drive of said disk drive array is available for storage of said plurality of subchunks of said plurality of titles.

4. The virtual file system of claim 1, further comprising:

a user process, executed on a storage processor node, which submits a title request for a selected title to said virtual file manager, which receives a corresponding directory entry for said selected title, and which submits a subchunk read request for each subchunk location entry in said corresponding directory entry;

wherein each subchunk read request is sent to a storage processor node identified by a storage processor node identifier in a corresponding subchunk location entry in said corresponding directory entry, and wherein each subchunk read request includes a destination node identifier, said disk drive identifier and said logical address; and

wherein said virtual file manager retrieves said corresponding directory entry for said selected title and forwards said corresponding directory entry to said user process in response to said title request.

5. The virtual file system of claim 4, further comprising a transfer process, executed on a storage processor node, which receives a subchunk read request, which requests a subchunk using said logical address to locate said requested subchunk from a local disk drive identified by said disk drive identifier, and which forwards a retrieved subchunk to a storage processor node identified by said destination node identifier.

6. The virtual file system of claim 4, wherein each title is subdivided into a plurality of data chunks, each said data chunk comprising a plurality of subchunks collectively comprising redundant data for each data chunk, and wherein said user process is operable to construct any data chunk from one less than all of said plurality of subchunks comprising said any data chunk.

7. The virtual file system of claim 6, wherein said disk drive array is divided into a plurality of redundant array groups, wherein each redundant array group comprises a plurality of disk drives distributed among a plurality of storage processor nodes, and wherein said plurality of subchunks of each data chunk are distributed among disk drives of a corresponding redundant array group.

8. The virtual file system of claim 7, wherein said user process is operable to reconstruct any stored title in the event of any one of:

a failure of any one disk drive;

a failure of any one disk drive of each of said plurality of redundant array groups; and

a failure of any one of said plurality of storage processor nodes.

9. The virtual file system of claim 8, wherein said user process is operable to reconstruct a missing subchunk of a data chunk from remaining subchunks of said data chunk, and is operable to return said reconstructed missing subchunk to a storage processor node that would otherwise have sourced said missing subchunk.

10. The virtual file system of claim 9, wherein in the event that failure of said storage processor node that would otherwise have sourced said missing subchunk is replaced by a replacement storage processor node, said replacement storage processor node re-stores missing and new title data by storing received subchunks including returned and reconstructed subchunks.

11. The virtual file system of claim 9, further comprising a cache memory, coupled to said storage processor node that would otherwise have sourced said missing subchunk, that temporarily stores received subchunks including returned and reconstructed subchunks for transfer to a replacement disk drive of a failed disk drive.

12. The virtual file system of claim 1, wherein each subchunk is stored in a block of a disk drive identified by said logical address, wherein said logical address comprises a logical block address.

13. The virtual file system of claim 1, wherein said virtual file manager manages title storage in which each title is subdivided into a plurality of data chunks, each data chunk comprising a plurality of subchunks incorporating redundant data for each data chunk.

14. The virtual file system of claim 13, wherein said disk drive array is divided into a plurality of redundant array groups, wherein each redundant array group comprises a plurality of disk drives distributed among a plurality of storage processor nodes, and wherein said plurality of subchunks of each data chunk are distributed among disk drives of a corresponding redundant array group.

15. The virtual file system of claim 14, further comprising:

a replacement disk drive with a plurality of missing subchunks coupled to a first storage processor node;

said virtual file manager preparing a disk repair directory entry listing each missing subchunk along with its corresponding parity subchunks comprising a data chunk and forwarding said disk repair directory entry to said first storage processor node; and

a repair process, executed on said first storage processor node, which submits a subchunk read request for each parity subchunk listed in said disk repair directory entry that corresponds to each missing subchunk, which reconstructs each missing subchunk using received corresponding parity subchunks, and which stores reconstructed subchunks onto said replacement disk drive.

16. The virtual file system of claim 15, further comprising:

a spare storage processor node;

a partially failed disk drive, replaced by said replacement disk drive, coupled to said spare storage processor node;

said virtual file manager forwarding said disk repair directory entry first to said spare storage processor node prior to sending to said first storage processor node; and

a salvage process, executed on said spare storage processor node, that employs checksum and locator to test validity of said missing subchunks stored on said partially failed disk drive, and that forwards valid subchunks read from said partially failed disk drive to said first storage processor node for storage on said replacement disk drive.

17. The virtual file system of claim 16, wherein said repair process discards received valid subchunks read from said partially failed disk drive in the event a corresponding missing subchunk has already been reconstructed and stored on said replacement disk drive.

18. The virtual file system of claim 14, wherein said disk drive array comprises a predetermined number of disk drives, and wherein said virtual file manager operates to distribute said plurality of data chunks in an even manner among said plurality of redundant array groups.

19. The virtual file system of claim 18, wherein said virtual file manager performs a re-striping process to re-distribute said plurality data chunks to maintain even distribution of data in response to a change of said predetermined number of disk drives.

20. The virtual file system of claim 19, wherein said re-striping process is performed as a background task.

21. The virtual file system of claim 19, wherein said virtual file manager performs said re-striping process to redistribute said plurality of data chunks among new disk drives in said disk drive array to maintain even distribution of data upon detecting an increase of said predetermined number of disk drives of said disk drive array.

22. The virtual file system of claim 19, wherein said virtual file manager detects a request to remove specified disk drives of said disk drive array, performs said re-striping process to redistribute said plurality of data chunks to maintain even distribution of data in remaining disk drives, and de-allocates said specified disk drives.

23. The virtual file system of claim 1, wherein said at least one management node comprises a mirror management node which executes a mirror virtual file manager which mirrors operation of said virtual file manager.

24. The virtual file system of claim 1, wherein said virtual file manager maintains a pool of pre-allocated directory entries, each comprising a list of available subchunk location entries.

25. The virtual file system of claim 24, wherein a number of said pool of pre-allocated directory entries is based on performance and site usage profile.